• Nenhum resultado encontrado

Martim Ribeiro da Cunha Manoel, nr. 431

N/A
N/A
Protected

Academic year: 2019

Share "Martim Ribeiro da Cunha Manoel, nr. 431"

Copied!
31
0
0

Texto

(1)

A Work Project, presented as part of the requirements for the Award of a Masters Degree in Management from the Faculdade de Economia da Universidade Nova de Lisboa.

The Dilution Effect: The Influence of Expertise and

Abstraction on Consumers’ Judgments of Products

Martim Ribeiro da Cunha Manoel, nr. 431

A Project carried out on the Consumer and Managerial Decision Making course, with

the supervision of:

Professor Els De Wilde

(2)

Abstract

Social studies suggest that adding obviously irrelevant product information to

diagnostic information mitigates consumers’ judgments of products, a concept named

dilution effect. This study shows that expert consumers are not only less subject to the

dilution effect than novice consumers but it also suggests that experts are in fact

shielded against the effects of irrelevant information. In addition, this project was also

able to demonstrate that irrelevant information may be positive for brands to

communicate when the target is not an expert in the product category and the irrelevant

information is somewhat abstract and vague. As irrelevant information becomes more

concrete, consumers with low expertise will be subject to the dilution effect whereas

experts will remain unaffected.

(3)

Contents

1. Introduction...3

2. Literature Review...5

2.1. The Dilution Effect...5

2.1.1. Averaging...6

2.1.2. Conversational Norms...7

2.1.3. Representativeness...7

2.1.4. Biased Hypothesis Testing...8

2.2. The Concept of Expertise and the Dilution Effect……….9

2.3. The Concept of Abstraction and the Dilution Effect……….…...10

3. Experiment...13

3.1 Method...13

3.2 Results...17

3.3 Discussion...19

4. Conclusion...20

5. Limitations and Future Research...21

6. References...22

(4)

1. Introduction

How can irrelevant information play a role in consumers’ decision processes? If

the information is irrelevant for a choice, should not it be discarded? Well, some studies

have shown that irrelevant information may indeed influence the decision making

process.

When consumers evaluate products or services, they often search for diagnostic

information on specific product benefits; information that helps them classify if a

product will deliver a desired benefit or not. Naturally, when searching for diagnostic

information, consumers will also be exposed to irrelevant information; information that

they perceive as clearly uninformative about the desired benefit they are looking for.

Normatively, the presence of irrelevant information should not change consumers'

assessment of the product's ability to deliver the desired benefit (Meyvis and

Janiszewski 2002). However, consumers do not always behave rationally. Studies on

social judgment have demonstrated that adding obviously irrelevant information to

diagnostic information may lead to less extreme judgments (e.g., De Dreu, Yzerbyt, and

Leyens 1995; Fein and Hilton 1992; Nisbett, Zukier, and Lemley 1981). This so-called

dilution effect suggests that obviously irrelevant product information may weaken

consumers’ beliefs in a product's ability to deliver a desired benefit (Meyvis and

Janiszewski 2002).

Contrarily, Carpenter, Glazer and Nakamoto (1994) argue that many brands also

successfully differentiate on an attribute that appears valuable but, on closer

examination, is irrelevant to creating the implied benefit. Despite not being obviously

irrelevant, it is still irrelevant for the benefit it proposes. For example, Alberto Culver

(5)

advertising it with the slogan "We put silk in a bottle" to suggest a user's hair will be

silky. However, a company spokesman conceded that silk "doesn't really do anything

for hair" (Adweek 1986). In the words of Carpenter, Glazer and Nakamoto (1994),

“consumers apparently value these differentiating attributes even though they are, in one

sense, irrelevant. Adding an irrelevant attribute to a brand changes the structure of the

decision that consumers face, especially if the differentiating attribute is difficult to

evaluate, such as silk in shampoo. As a result, consumers may infer the attribute's value

and, in some cases, conclude that it is valuable”.

Two examples have been shown in which irrelevant information has an influence

in consumers’ judgments regarding the ability of a certain product to deliver a desired

benefit. But at this point two questions emerge: (1) does this happen to consumers with

different levels of expertise? And (2) What about the type of information displayed,

does it play a role?

To answer the first question, this study tries to assess if the level of expertise (or

experience) may mitigate the effects of irrelevant information. The literature in

psychology indicates that the dilution effect may diminish with experience, possibly

because more-experienced decision makers have highly developed knowledge structures

(organization of knowledge in memory) that enable them to focus on relevant evidence

(Patel and Groen 1986; Lesgold et al. 1988).

In addition to expertise, this project also tries to evaluate whether the level of

abstraction of a product attribute mitigates (or not) the impact of irrelevant information.

Abstractness is typically defined as the inverse of how directly an attribute denotes

particular objects or events (Johnson 1984). Conversely, concrete attributes are those

(6)

object need to be inferred or computed from concrete attribute information, whereas

concrete attributes are directly associated with the object (Bettman and Sujan 1987). In

the example of “Alberto Natural Silk Shampoo”, silk was an irrelevant product attribute

but it was also difficult for consumers to evaluate. One of the propositions of this study

is that abstract information may have similar consequences to the ones reported in the

previous example.

Despite not being areas that have intensively been studied, there are already some

conclusions of how the level of expertise relates to dilution and how the level of

abstraction of product information may interact with trivial attributes. However,

combining both variables (expertise and abstraction) has not yet been studied and that is

the main objective of this project, to assess whether different levels of expertise

combined with different levels of abstraction generate more or less dilution in

consumers’ judgments of a product’s ability to deliver a desired benefit.

2. Literature Review: Dilution Effect, Expertise, and Abstraction

In this section of the project I will discuss the literature on the dilution effect, and

discuss the effects of expertise and abstraction on dilution.

2.1. The Dilution Effect

When irrelevant information is added to diagnostic information, consumers

generate less extreme judgments, a phenomenon also known as the dilution effect.

Several studies in the consumer behavior literature have demonstrated that objectively

irrelevant product information can influence consumer decisions (Meyvis and

Janiszewski 2002). Zukier and Jennings (1983) reported a good example of this

(7)

to find a man guilty of murdering his aunt when irrelevant information (e.g., “The

defendant is of average height and vision”) was added to diagnostic information (e.g.,

“He was known to have argued with his aunt”).

In an attempt to explain why this phenomenon happens, four main theories have

been used: averaging models, conversational norms, representativeness heuristic and

biased hypothesis testing.

2.1.1. Averaging

The first explanation for the dilution effect is called averaging and it defends that

by adding irrelevant information to supportive information, the impact of the latter is

somehow reduced.

Averaging is the most popular account of the dilution effect within the nonsocial

judgment literature and can be classified into two groups of models (Birnbaum and

Mellers 1983; Lichtenstein, Earle and Slovic 1975; Shanteau 1975; Troutman and

Shanteau 1977). Defenders of the first group, (e.g., Anderson 1971, 1974; Birnbaum

and Mellers 1983) argue that if each piece of information added has a nonzero weight,

and assuming that each message always accounts for 100%, the addition of irrelevant

information will always have some weight that will force the reduction in the weight of

already existing attributes. Researchers in favor of a different explanation (e.g.,

Lichtenstein et al. 1975; Lopes 1987; Shanteau 1975) assume that people make separate

predictions regarding each piece of information and then average their predictions

(8)

2.1.2. Conversational Norms

A different explanation for the dilution effect comes from social judgment

literature. It is argued by authors such as Schwarz et al. (1991); Slugoski and Wilson

(1998); Tetlock, Lerner, and Boettger (1996) that the dilution effect is merely a result

observed in experiments due to erroneous inferences made by people that rely on

conversational norms. One of the conversational norms that is assumed to be respected

in such experiments is that “all information is relevant for the goal of the conversation”

(Grice 1975), also called the principle of relevance by Sperber and Wilson (1986). “If

the experimenter provides this information it must be relevant” is the reason why the

principle of relevance is violated, leading participants in dilution studies to wrong

conclusions.

2.1.3. Representativeness

Still within the social judgment theory, the most popular explanation for the

dilution effect for many years has been the representativeness heuristic (Fein and Hilton

1992; Hilton and Fein 1989; Locksley, Hepburn, and Ortiz 1982; Nisbett et al. 1981;

Tetlock and Boettger 1989; Zukier 1982). According to Kahneman and Tversky (1972,

1973) “the representativeness heuristic is a strategy by which subjects use the similarity

between the available information about the individual and the to-be-predicted behavior

to estimate the probability that the individual will display the behavior”. In other words,

consumers have in mind a product/brand they recognize as being representative (e.g.,

Gillette may be seen as representative of the razor blade category) and when they

compare a new product against the one they know, if the new product has characteristics

(irrelevant) that the other does not have, it will be seen as different, thus less

(9)

example of this explanation was given in a study carried by Nisbett et al. (1981) in

which subjects considered a man with a drinking problem and two fingers missing on

his left hand less representative of the stereotypical child abuser than a man with a

drinking problem only. Therefore, even though the fingers missing on one of his hands

was irrelevant, it made the subject less representative.

2.1.4. Biased Hypothesis Testing

One last and more recent explanation that helps understanding the dilution effect

even when the other theories seem not to hold is called biased hypothesis testing. It

argues that people often consider the implications of evidence for the focal hypothesis

but ignore the implications of the evidence for the alternative hypothesis (e.g., Trope

and Liberman 1996). Fischhoff and Beyth-Marom (1983) suggest that, “by selectively

focusing on a single hypothesis, people may decrease their confidence in the hypothesis

when encountering unlikely irrelevant information”.

Meyvis and Janiszewski (2002) argue that consumers will selectively search for

information that suggests that the product will deliver the benefit, and that they will

classify product information with respect to this goal search in two possible ways: (1)

when information is classified as confirming, it strengthens consumers’ beliefs that the

product will deliver the benefit; (2) when information is classified as not confirming, it

weakens consumers’ beliefs that the product will deliver the benefit. In addition, when

consumers encounter irrelevant information, they classify it as not confirming, thus

weakening their belief in the product’s ability to deliver the desired benefit, without

taking into account that the information does not confirm the alternative hypothesis

(10)

2.2. The Concept of Expertise and the Dilution Effect

As it was shown earlier in this project, several efforts have been made to explain

the dilution effect and some studies have examined the interaction between expertise

and dilution. Still, most of these studies do not include evaluations of products by

consumers nor the interaction with different types of information, which are the main

objectives of this project.

Intuitively, one would expect that as the level of expertise of a person in a certain

area increases, the dilution effect decreases since experts will know what is relevant and

irrelevant for taking a decision. This conclusion was supported by Shelton (1999) in a

study regarding audit tasks. Shelton divided participants in her study into experts (3 or

more years in an auditing function) and novices (less than 3 years in an auditing

function) and provided the same material to both groups, in order for them to perform

an audit task. The data provided was composed by relevant and irrelevant financial

information. Participants in the high expertise condition were not only less subject to

dilution than novices but they also seemed to be shielded against the effects of

irrelevant information. The same researcher attempted possible explanations for this:

“Findings in psychology indicate that highly developed knowledge structures are

acquired through experience. Such knowledge structures could enable experienced

decision makers to focus on relevant information. This implies that the dilution effects

observed in prior studies may have been weaker had judgments been made by

more-experienced subjects”. Similarly, Lesgold et al. (1988) argue that experts are able to

tune their perceptions of critical areas to specific cases and discount irrelevant

information which novices use in their decisions. The same authors and Shanteau

(11)

experts tend to ignore irrelevant information while novices seem unable to do so even

when the irrelevancy of the information is recognized prior to making the decision.

Despite being a topic somewhat studied, it is still of use to assess whether experts

are in fact less subject than novices to the dilution effect when the category is

product-related. Still, the major contribution that this study proposes is not the simple

relationship of expertise and dilution but the interaction of expertise, the type of

information (abstraction) and dilution, as it will be mentioned in the following section.

2.3. The Concept of Abstraction and the Dilution Effect

Following from the literature on the dilution effect, there is evidence that

irrelevant information does in fact influence consumers’ opinions. However, different

types of irrelevant information may lead to different results. What is the impact of the

level of abstraction of irrelevant information on consumers’ judgments of products?

Many products have differentiating features with benefits that are unclear to

consumers. Despite uncertainty surrounding the benefits of ambiguous features, several

researchers observe that consumers may incorporate these features into their

decision-making (e.g., Bronziarczyk and Gershoff 1997; Brown and Carpenter 2000; Carpenter,

Glazer and Nakamoto 1994; Meyvis and Janiszewski 2002; Mukherjee and Hoyer

2001). As it was said earlier in this study, abstract attributes of an object derive from

concrete attribute information. In other words, the concrete attributes are directly

associated with an object and can be classified as a subcategory of abstract information.

The abstract-concrete distinction is similar to the superordinate-basic level distinction

drawn in the categorization literature (Mervis and Rosch 1981; Rosch et al. 1976).

“Basic level attributes are those that are naturally associated with the stimulus object

(12)

general or inconclusive attributes that subsume the basic level attribute (e.g., fun,

enjoyment)” (Bettman and Sujan 1987). Following the same line of reasoning,

“French-cut” green beans was used by Miljkovic, Gong and Lehrke (2009) as an example of

food product related trivial attributes that do not reveal any useful information about the

product. What I propose here is that abstract product attributes share a similar

connotation with ambiguous/difficult to evaluate product attributes.

Marketing experts (e.g., Aaker 1991; Porter 1985) have extensively documented

their opinions and studies about product differentiation and the major conclusion drawn

is that only attributes that are relevant, meaningful and valuable generate successful

product differentiation (Carpenter, Glazer and Nakamoto 1994). However, there is

evidence that using trivial but ambiguous attributes may also generate successful

product differentiation. For instance, Hoch and Ha (1986) observed that a

nondiagnostic, ambiguous product experience can increase the perceived quality of an

advertised brand. Similarly, Carpenter Glazer and Nakamoto (1994) reported that a

brand with a distinguishing but irrelevant attribute received a higher preference rating

than the same brand without the attribute. The same authors defend that trivial attributes

affect choice through their uniqueness in the choice set. In other words, by having an

attribute that is different from the ones used by competitors, it is argued that successful

differentiation occurs since it makes a brand unique relatively to others. Studies of

causal inference by Einhorn and Hogarth (1986) support the theory that the uniqueness

of the irrelevant attributes can lead to positive valuations of the differentiated brand. In

addition to these researchers, McGill (1989) also shows that causality is attributed more

often to distinctive rather than common attributes, suggesting that buyers may infer that

(13)

This project proposes that despite being irrelevant, there will be a difference in

consumers’ judgments of products when the level of abstraction of the attributes

changes. As it was stated earlier in this section, the prediction is that abstract irrelevant

information, in the context of dilution studies, will have results consistent with the ones

found in studies about ambiguous product attributes. The literature has shown that

attributes that are abstract, ambiguous and vague (e.g., “Easy to use microwave”,

“Funny recorder”, and so on) can strengthen consumers’ judgments of products because

the value that they bring is uncertain. Thus, they tend to believe it is actually better.

Contrarily, as the irrelevant attributes become more concrete, consumers will be able to

identify and classify the information as irrelevant but dilution will occur. In addition, as

it was stated in the previous section, consumers with high levels of expertise are not

expected to be as influenced by irrelevant information as consumers with low levels of

expertise, regardless of the level of abstraction of the information displayed. As a result,

I hypothesize that:

H1: When evaluating a product’s ability to deliver a desired benefit, expert consumers

will be less influenced by irrelevant information than novice consumers, regardless of

the level of abstraction of the information.

H2: If a brand adds irrelevant pieces of information or attributes to a product that are

abstract and vague, novice consumers’ beliefs that the product will deliver the benefit

will be strengthened.

H3: Analogously, if a brand adds irrelevant pieces of information or attributes to a

product that are concrete, novice consumers’ beliefs of a product’s ability to deliver a

(14)

3. Experiment

This experiment has two main objectives: (1) evaluate whether experts are less

subject (or not) than novices to the influence of irrelevant information when judging

products and (2) study how the level of abstractness relates to the dilution effect for

novices and experts.

3.1. Method

Subjects and Design

Subjects were 60 randomly selected participants from both genders and various

professional occupations that agreed to answer to some questions to help collecting data

for a master thesis.

The design was a 3 (type of information) x 8 (product replicates) x 2

(expert/novice consumer) mixed design. Each subject was presented with descriptions

of products from two categories: (1) snowboarding and (2) consumer goods for body

care. In total, each subject received a description of eight products, four from each

category. For each of the product replicates, subjects were randomly assigned to either

the baseline condition or the other two treatment conditions. Finally, all subjects filled

out an expertise scale for the two categories.

Procedure

The entire experiment was conducted online. Subjects were informed that they

would receive information about eight different products and that they would have to

indicate whether the product would deliver a particular benefit. Subjects were told that

the information they would receive “may or may not be helpful for the decision that you

(15)

(e.g., “You are looking for toothpaste that fights cavities”). Below this sentence, there

was the first piece of information, which was always the supportive attribute (e.g., “The

majority of dentists recommend it”). In the baseline condition, subjects did not receive

any additional product information. However, in the first treatment condition, the

supportive information was presented simultaneously with three additional pieces of

abstract irrelevant information (e.g., “Provides alpine freshness”, “Exotic flavor”,

“Excellent durability”). Similarly, in the second treatment condition, the supportive

information was presented simultaneously with three pieces of concrete irrelevant

information (e.g., “Composed by 70% of mint leaves”, “Composed by 40% of

Citrinus”, “Comes in 40cl tubes”). Afterwards, subjects were asked to indicate their

belief that the product would deliver the benefit (e.g., “Does this toothpaste fight

cavities?”). Responses were made on a nine-point scale anchored by 1 = “Definitely

does NOT fight cavities” and 9 = “Definitely does fight cavities”. The subjects then

received the remaining seven replicates.

To measure expertise, I followed a similar approach to the one used by Sujan

(1985), in which he measured expertise about photographic cameras by having

participants complete a 15-question multiple choice test to measure objective

knowledge about cameras. Since it would be difficult to have a 15-question multiple

choice test for each product, I selected eight products that belong to two categories: (1)

four snowboarding products and (2) four consumer goods for body care.

Finally, when product evaluations were completed, subjects were asked to

complete two 15-question multiple choice tests, one about the snowboarding sport (e.g.,

“A regular snowboarder is...”, “The main snowboarding styles are...”, “A grind 50/50

(16)

it is recommended to use lotions with...”, “Shea butter is especially useful for...”, “An

exfoliating cream is...”), in order to assess who were the experts and novices in each

product category. Participants whose score to each test was below and above the median

were considered novices and experts, respectively. For a complete version of the two

expertise tests, please see appendices 1 and 2.

Stimuli

I first selected eight products from two product categories and corresponding

desired benefits. Snowboarding products and corresponding desired benefits: boots

(comfortable), board (easy to maneuver), bindings (easy to strap), goggles (anti-fog).

Consumer goods for body care and corresponding desired benefits: toothpaste (fights

cavities), shower gel (anti skin irritation), shampoo (anti-dandruff), lotion (rejuvenates

the skin). Following a similar approach to the one used by Meyvis and Janiszewski

(2002), a pretest (n = 31) was conducted to select three irrelevant attributes and one

supportive attribute for each replicate. The pretest listed a wide range of facts for each

product. Subjects were asked to allocate these facts to one of three categories: “suggests

[benefit]”, “suggests not [benefit]”, “is not helpful for my decision”. The 48 irrelevant

facts selected for the experiment were classified as “not helpful” by an average of 88%

of pretest subjects, as supportive of the benefit by 9% of the subjects, and as

counterdiagnostic by 3% of the subjects. The irrelevant information included mainly

package information (e.g., toothpaste that comes in 30cl tubes) and product attributes

(e.g., snowboard goggles that can be ordered online). The eight supportive facts were

classified as suggesting the benefit by an average of 92% of pretest subjects.

A second pretest was conducted to examine the possibility that the facts judged as

(17)

description. Thirty one subjects were presented with the full product descriptions and

asked to indicate the relevance of each piece of information. The 48 irrelevant facts

were classified as nondiagnostic by an average of 86% of the subjects, as diagnostic of

the benefit by 9% of the subjects, and as counterdiagnostic by 5% of the subjects.

A third pretest examined the possibility that even though subjects indicated that

the irrelevant facts are not diagnostic, they may still use these facts to make inferences

about the desired benefit. For example, although a subject may classify a fact as

irrelevant, the fact may still be informative because it is positively or negatively

correlated with unstated facts that are relevant. To examine the possible direction of

such an effect, 20 subjects were presented with the irrelevant facts and were asked to

rate them on a six-point scale (ranging from 1 = “Will probably not [deliver the

benefit]” to 6 = “Will probably [deliver the benefit]”), thus forcing them to classify the

information as either diagnostic or counterdiagnostic. The abstract irrelevant facts were

classified as diagnostic by an average of 41% of the subjects and as counterdiagnostic

by an average of 59% of the subjects. The average rating for the abstract irrelevant facts

was 3.05, which was significantly lower than 3.50, the midpoint of the scale (t(480) =

-4.23, p < 0.01). Thus, if the abstract irrelevant information would indeed lead to

inferences about the benefit, these inferences would not support subjects’ beliefs in the

product benefit. The concrete irrelevant facts were classified as diagnostic of the benefit

by an average of 50% of the subjects and as counterdiagnostic by an average of 50% of

the subjects. The average rating for the concrete irrelevant facts was 3.29, which is

slightly lower than 3.50 (t(480) = -2.13, p = 0.033). Thus, even though the difference is

(18)

the benefit, there is a slight tendency not to support subjects’ beliefs in the product

benefit.

In a final pretest, thirty subjects were asked to state whether each of the selected

48 irrelevant facts were abstract or concrete. Subjects were presented with the attributes

and classified them into to one of two categories: “This attribute is abstract” or “This

attribute is concrete”. The abstract irrelevant attributes were classified as abstract by an

average of 89% of the subjects and as concrete by an average of 11% of the subjects.

Similarly, the concrete irrelevant attributes were classified as concrete by an average of

87% of the subjects and as abstract by 13% of the subjects.

3.2. Results

It was expected that the addition of irrelevant information would generate stronger

effects in the beliefs in the product benefit for novices (low expertise) than for experts

(high expertise).

Table 1: Summary of the average rating per condition

The results confirmed what was predicted by H1, that experts are less influenced

(19)

possible to confirm that expert consumers showed small variations in both treatment

conditions relatively to the “Relevant” condition (∆ -0.09 in the “Relevant + Abstract

Irrelevant” condition and ∆ +0.34 in the “Relevant + Concrete Irrelevant” condition).

The average expert ratings in the 3 conditions are statistically non-significant ((t(160) =

0.34 , p = 0.74 ; t(160) = 0.96 , p = 0.34 ; t(160) = 0.68 , p = 0.5)1. On the contrary, the

upper row of the same table shows that there are differences in the average novice rating

in the 3 conditions (t(160) = -2.67 , p < 0.01 ; t(160) = -5.12 , p < 0.01 ; t(160) = -2.37 ,

p = 0.02).

Regarding H2, which predicted that novice consumers’ beliefs in product benefit

would be strengthened when abstract product information was added to relevant

information, results also confirmed what was proposed by the theory. The average

variations for consumers with low expertise were significantly higher, with a variation

of ∆ +0.77 in the “Relevant + Abstract Irrelevant” condition and average variations of ∆

-0.68 in the “Relevant + Concrete Irrelevant” condition. Supporting these results, the

average rating from novice consumers in the “Relevant + Abstract Irrelevant” condition

( = 6.38) was higher than the average rating from novice consumers in

the “Relevant” condition ( = 5.61), a difference that is statistically

relevant (t(160) = -2.67 , p < 0.01).

Finally, in what concerns H3, the results also confirmed what was proposed by the

theory. It was suggested that concrete irrelevant information, when added to relevant

information, would dilute novice consumers’ beliefs in the product benefit and this

prediction was confirmed. The average rating in the “Relevant + Concrete Irrelevant”

condition ( = 4.93) was lower than the average rating in the “Relevant”

(20)

condition ( = 5.61), which is statistically significant (t(160) = -2.37 , p =

0.02).

Interestingly enough, there is also a statistically significant difference between the

average rating from novice and expert consumers in both conditions when compared to

one another (t(160) = 2.42 , p = 0.02). In the “Relevant + Abstract Irrelevant” condition,

consumers with low expertise are positively influenced when the abstraction level is

higher ( = 6.38), compared to consumers with high levels of expertise

receiving the same information ( = 5.69). Contrarily, in the “Relevant +

Concrete Irrelevant” condition, the average rating from novice consumers

( = 4.93) is statistically lower (t(160) = 2.14 , p < 0.01) than the average

rating from expert consumers ( = 5.95).

3.3. Discussion

Research on social judgment suggests that irrelevant information does have an

impact on consumers’ product perceptions. The main argument is that irrelevant

information weakens consumers’ beliefs in the product benefit. This study examined the

effects of abstraction and expertise on dilution.

This study suggests that irrelevant information does in fact influence consumers’

product perceptions when their knowledge of the product category is low, but not when

it is high. When consumers are considered experts in a product category, are

knowledgeable about it and/or have used this type of products in the past, irrelevant

information does not appear to influence their opinion of products, they are not affected.

Contrarily, when consumers where not familiar with the product category, have

(21)

a role in their perception of products. Even more interesting than the fact that the group

of participants classified as novices was more influenced than the group of experts, is

the direction of the effects. When irrelevant information with higher levels of

abstraction was added to relevant product information, subjects’ perception of products

improved substantially. Contrarily, adding irrelevant information with lower levels of

abstraction (concrete irrelevant product information) revealed the exact opposite effect

on consumers with little knowledge of a product category.

One possible explanation for this is that consumers with no experience in a

product category tend to be positively influenced when they do not know the true value

of the information and as the attributes become easier for them to evaluate and

understand, they are able to classify them as irrelevant. Still, they are not able to discard

them; novice consumers fall in the “trap” called the dilution effect.

4. Conclusion

In this project I examined how the level of expertise in a product category and

how the level of abstraction of irrelevant product information affect consumers’

perception of products. I conclude that when a consumer is an expert in a certain area,

he/she will not be influenced by irrelevant information of any type. Contrarily, when a

consumer is not an expert in a certain area he/she will be affected by irrelevant

information. What this study suggests is that the direction of the effects will depend on

the level of abstraction of the irrelevant information. When irrelevant information is

more abstract, it will positively influence consumers’ perception of products but when

the irrelevant information is more concrete, the products’ ability to deliver the desired

benefits will be considered less extreme by consumers, meaning that the “classic”

(22)

It is common for brands to change their communication programs according to the

type of target consumers they are trying to reach and this study draws conclusions that

brands may adopt. If by any reason a brand is not certain of the amount or type of

information to display when communicating a product, in a campaign or in the product

packaging, this study suggests that more abstract and vague information will generate

better perceptions by consumers, when their knowledge about the product category is

little. Ceteris paribus, using information easy to evaluate and more concrete may

produce the exact opposite effect to the one intended when setting up a communication

program. In other words, instead of displaying more information in order that

consumers generate more favorable judgments of a product or brand, consumers will

show the opposite action, their judgments that a product will deliver a desired benefit

will be mitigated.

5. Limitations and Future Research

There are some areas in this project that could be improved in the future. The

difference between an expert and a novice consumer can be considered as non-rigorous

since there is the possibility that some of the respondents that were classified as experts

did “pass the test” by pure luck.

Only two product categories were used, leaving some uncertainty if the

conclusions are still valid and hold in other product categories and services.

In addition to exploring other products and services, as the level of abstraction

suggested some influence, there may be many more divisions that can be made in the

type of information variable that may provide further insights and allow a better

(23)

6. References

Aaker, D. A. (1991), “Managing Brand Equity,” New York: The Free Press.

Adweek (1986), "Silk in a Bottle" (May 19), 18.

Anderson, N. H. (1971), "Integration Theory and Attitude Change," Psychological Review, Vol 78(3), (May), 171-206.

Anderson, N. H. (1974), "Information Integration Theory: A Brief Survey," Contemporary Developments in Mathematical Psychology. Vol. 1. Learning, Memory and Thinking, ed. David H. Krantz et al.. 236-270.

Bettman, J. R. and M. Sujan (1987), "Effects of Framing on Evaluation of Comparable and Noncomparable Alternatives by Expert and Novice Consumers," Journal of Consumer Research, 14(September), 141-154.

Birnbaum, M. H. and B. A. Mellers (1983), "Bayesian Inference: Combining Base Rates with Opinions of Sources Who Vary in Credibility," Journal of Personality and Social Psychology, 45 (October), 792-804.

Broniarczyk, S. M. and A. D. Gershoff (1997), “Meaningless differentiation revisited,” Advances in Consumer Research, 24, 223-228.

Brown, C. L. and G. S. Carpenter (2000), “Why is the trivial important? A reason-based account for the effects of trivial attributes on choice,” Journal of Consumer Research, 26, 372-385.

Carpenter, G. S., R. Glazer, and K. Nakamoto (1994), “Meaningful Brands From Meaningless Differentiation: The Dependence on Irrelevant Attributes,” Journal of Marketing Research, Vol. XXXI (August 1994), 339-350.

De Dreu, C. K. W., V. Y. Yzerbyt and J. P. Leyens (1995), "Dilution of Stereotype-Based Cooperation in Mixed-Motive Interdependence,” Journal of Experimental SocialPsychology, 31 (November), 575-593.

(24)

Fein, S. and J. L. Hilton (1992), "Attitudes toward Groups and Behavioral Intentions toward Individual Group Members: The Impact of Nondiagnostic Information," Journal of ExperimentalSocial Psychology, 28 (March). 101-124.

Fischhoff, B. and R. Beyth-Marom (1983), "Hypothesis Evaluation from a Bayesian Perspective," Psychological Review, 90 (July). 239-260.

Grice, H. P. (I975), "Logic and Conversation," in Syntax: and Semantics. Vol. 3, Speech Act.\. ed. Peter Cole and Jerry L. Morgan, New York: Academic Press. 41-58.

Hilton, J. L. and S. Fein (1989), "The Role of Typical Diagnosticity in Stereotype-Based Judgments," Journal of Personality and Social Psychology, 57 (August), 201-21 1.

Hoch, S. J., Y. W. Ha (1986), “Consumer Learning: Advertising and the ambiguity of product experience,” Journal of Consumer Research, 13: 221-223.

Johnson, M. D. (1984), “Consumer Choice Strategies for Comparing Noncomparable Alternatives,” Journal of Consumer Research, 11 (December), 741-753.

Kahneman, D. and A. Tversky (1972), "Subjective Probability: A Judgment of Representativeness," Cognitive Psychology, 3 (July), 430-454.

Kahneman, D. and A. Tversky (1973), "On the Psychology of Prediction," Psychological Review, 80 (July), 237-251.

Lesgold, A., H. Rubinson, P. Feltovich, R. Glaser, D. Klopfer, and Y. Wang (1988), “Expertise in a complex skill: Diagnosing x-ray pictures,“ The Nature of

Expertise, edited by M. T. H. Chi, R. Glaser, and M. J. Farr, 3-11-342. Hillsdale, NJ: Erlbaum.

(25)

Locksley, Anne, Christine Hepburn, and Vilma Ortiz (1982), "Social Stereotypes and Judgments of Individuals: An Instance of the Base-Rate Fallacy," Journal of Experimental Social Psychology, 18 (January), 23-42.

Lopes, L. L. (1987), "Procedural Debiasing," Acta Psychologica, 64 (February), 167-185.

McGill, A. L. (1989), “Context in Judgments of Causation,” Journal of Personality and Social Psychology, 57, 189-200.

Mervis, C. B. and E. Rosch (1981), “Categorization of Natural Objects,” Annual Review of Psychology, 32, 89-115.

Meyvis, T. and C. Janiszewski (2002), “Consumers’ Beliefs about Product Benefits: The Effect of Obviously Irrelevant Product Information,” Journal of Consumer Research, Vol. 28: 618-635.

Miljkovic, D., J. Gong, and L. Lehrke (2009), “The Effects of Trivial Attributes on Choice of Food Products,” Agricultural and Resource Economics Review, 38/2 (October 2009), 142-152.

Mukherjee, A. and W. D. Hoyer (2001), “The Effect of Novel Attributes on Product Evaluation,” Journal of Consumer Research, 28, 462-472.

Nisbett, R. E., H. Zukier, and R. E. Lemley (1981), "The Dilution Effect: Nondiagnostic Information Weakens the Implications of Diagnostic Information," Cognitive Psychology, 13 (April), 248-277.

Patel, V. L., and G. L. Groen, (1986), “Knowledge based solution strategies in medical reasoning,” Cognitive Science, 10: 91-116.

Porter, M. E. (1985), “Competitive Advantage,” New York: The Free Press Review. 78 (May), 171-206.

(26)

Schwarz, N., F. Strack, D. Hilton, and G. Naderer (1991), "Base Rates.

Representativeness and the Logic of Conversation: The Contextual Relevance of Irrelevant Information," Social Cognition, 9 (Spring), 67-84.

Shanteau, J. (1975), "Averaging versus Multiplying Combination Rules of Inference Judgment," Acta Psychologica, 39 (February), 83-89.

Shanteau, J. (1993), “Discussion of expertise in auditing,” Auditing: A Journal of Practice & Theory, 12 (Supplement), 51-56.

Shelton, S. W. (1999), “The Effect of Experience on the Use of Irrelevant Evidence in Auditor Judgment,” The Accounting Review, (April), Vol. 74: 217-224.

Slovic, P. (1972), “From Shakespeare to Simon: Speculation – and Some Evidence – About Man’s Ability to Process Information,” Oregon Research Institute Research Monograph, 12 (3).

Slugoski, B. R. and A. E. Wilson (1998), "Contribution of Conversation Skills to the Production of Judgmental Errors," European Journal of Social Psychology, 28 (July), 575-601.

Sperber, D. and D. Wilson (1986), “Relevance: Communication and Cognition,” Cambridge, MA: Harvard University Press.

Sujan, M. (1985), “Consumer Knowledge: Effects on Evaluation Strategies Mediating Consumer Judgments,” Journal of Consumer Research, 12 (June), 31-46.

Tetlock, P. E. and R. Boettger (1989), "Accountability: A Social Magnifier of the Dilution Effect," Journal of Personality and Social Psychology, 57 (September), 288-398.

(27)

Trope, Y. and A. Liberman (1996), "Social Hypothesis Testing: Cognitive and

Motivational Mechanisms," in Social Psychology: Handbook of Basic Principles, ed. Edward Tory.

Troutman. C. M. and J. Shanteau (1977), "Inferences Based on Nondiagnostic

Information,” Organizational Behavior and Human Performance, 19 (June), 43-55.

Zukier, H. (1982), "The Dilution Effect: The Role of the Correlation and the Dispersion of Predictor Variables in the Use of Nondiagnostic information,” Journal of Personality and Social Psychology, 43 (December), 1163-1174.

(28)

7. Appendices

Appendix 1: Snowboarding Expertise Test

Em seguida estarão 15 perguntas de escolha múltipla sobre Snowboard e a ideia é responderes o melhor que souberes.

1) Um “air to fakie” é…:

a) Um tipo de prancha de Snowboard

b) Um tipo de botas de Snowboard

c) Uma manobra

d) Andar com o pé contrário à frente

2) Os principais estilos de Snowboard são…: a) Freestyle, carving, switching e free ride

b) Free ride e carving

c) Freestyle, free ride e carving

d) Freestyle and free ride, carving e splitboarding

3) Um Snowboarder Regular é geramlente usado para designar…:

a) Alguém que prefere andar com o pé esquerdo à frente

b) Alguém que faz Snowboard regularmente

c) Alguém que prefere andar com o pé direito à frente

d) Alguém que anda sem cometer praticamente nenhum erro ou que raramente cai

4) Um Grind 50/50 é…:

a) O melhor capacete inventado até hoje e que cumpre todos os 50 parâmetros de segurança

b) Um estilo de snowboard

c) Um salto em que só se cumpre metade do obstáculo

d) Uma manobra que se salta para cima duma caixa ou corrimão e se vai sempre em frente

5) Snowboarders inexperientes devem usar pranchas…:

a) Menos flexíveis para fazerem movimentos mais correctos

b) Mais flexíveis para ser mais fácil executar movimentos

c) Mais flexíveis e compridas para obrigar a curvas mais largas e menos repentinas

d) Mais flexíveis e mais curtas

6) Uma prancha para Freestyle é…: a) Composta por um Twin Tip

b) Maior do que a de Freeride para maior controlo nos saltos

c) Menos flexível que a de Freeride para aguentar o impacto nas caixas e corrimões

d) Mais estreita que a de Freeride

7) O tipo de botas de snowboard mais utilizadas são as…:

a) Soft Boots

b) Hard boots

c) Step-In Boots

d) Maneuver Boots

8) Flow-In é um tipo de…: a) Fixações

b) Manobra que se pode efectuar numa caixa

c) Casaco que liga às calças para oferecer maior protecção da neve e frio

d) Máscara para condições climatéricas extremas

9) Na montanha, a prioridade é sempre de quem…:

a) Vem da direita

b) Vem da esquerda

c) Vai atrás

d) Vai à frente

10) A posição dos pés recomendada para um freestyler é…:

a) Abertos como os ponteiros de um relógio às “10 para as duas”

b) O pé da frente está mais aberto que o de trás

c) O pé de trás está mais aberto que o da frente

d) Os dois pés virados para a frente da prancha

11) Andar em Fakie é…: a) Andar sempre em frente

b) Fazer curvas muito rápidas sem derrapar

c) O mesmo que andar em switch

d) Andar com o menor contacto possível da prancha com a neve

12) Quando se vai para o Snowpark, é recomendado levar as fixações…:

a) Ligeiramente mais apertadas para a prancha responder aos pequenos movimentos

b) Ligeiramente menos apertadas para maior liberdade dos pés

c) Step In

(29)

13) Burton é…:

a) Uma marca de material de Snowboard

b) Um forfait especial só para o Snowpark

c) Uma manobra de Snowboard efectuada num corrimão

d) Nenhuma das outras respostas

14) Para andar mais depressa é preciso…: a) Usar botas mais leves

b) Ter a prancha encerada

c) Ter a prancha limpa com um detergente específico

d) Ter um peso mais leve

15) Andar apenas com o Nose ou Tail da prancha em contacto com a neve é um truque chamado…:

a) Ollie

b) Air to Fakie

c) 50/50 Grind

(30)

Appendix 2: Consumer Goods Expertise Test

Em seguida estarão 15 perguntas de escolha múltipla sobre produtos para o corpo e a ideia é responderes o melhor que souberes.

1) Pseudofoliculite é também conhecida por…:

a) Vitamina que alisa o cabelo

b) Vitamina que hidrata a pele

c) Pêlos encravados

d) Um tratamento para a pele

2) Para evitar irritação da pele é recomendado usar cremes…:

a) Com extractos de Kiwi

b) Com ph baixo

c) Com um teor de álcool mais elevado

d) Com ph neutro

3) Uma pele oleosa também significa que…: a) A pele está desidratada

b) A pele está hidratada

c) Há maior propensão a ter cabelo oleoso

d) Nenhuma das restantes respostas está certa

4) Palmolive é uma marca que pertence à…: a) Unilever

b) Procter & Gamble

c) Colgate

d) L’Oreal

5) Um creme esfoliante é…:

a) Um creme que limpa as impurezas e células mortas da pele

b) Um creme que, quando aplicado antes do Verão, permite um bronzeado mais duradouro

c) Um creme amaciador para o cabelo para aplicar após o banho

d) Um creme que ajuda a eliminar a celulite

6) A manteiga de Karité é especialmente útil…: a) Em casos de queimaduras solares e alergias causadas pelo sol

b) Para combater a caspa

c) Quando se quer dar volume ao cabelo

d) Para potenciar o bronzeado

7) Para evitar cortes na pele, deve fazer-se a barba…:

a) Antes do banho pois a pele está mais dura e menos susceptível a cortes

b) Ao acordar pois a pele está relaxada do sono

c) Antes de ir dormir para poder descansar durante o sono

d) Depois do banho

8) O flúor de uma pasta de dentes…: a) Actua como agente branqueador

b) Não pode ser ingerido em grandes quantidades pois pode matar

c) Acumula-se nas gengivas, dando-lhes uma maior resistência

d) Não tem qualquer efeito, é apenas uma invenção das marcas dentífricas

9) Usar um creme protector solar com SPF 50 significa que…:

a) A pele só ficará queimada quando tiver sido exposta a uma quantidade de energia solar 50 vezes superior à quantidade de energia solar que normalmente a queimaria

b) Temos de repor uma nova camada de creme a cada 50 minutos

c) O creme tem quantidade suficiente para 50 pessoas de estatura média

d) A pele começa a ficar queimada passados 50 minutos da sua aplicação

10) Em termos concretos, uma pele fica bronzeada…:

a) Porque não temos cuidado suficiente com a pele

b) Porque geralmente o gel de banho que usamos tem um químico que escurece a pele quando esta entra em contacto com a radiação ultravioleta

c) Porque existe um aumento ou libertação do pigmento melanina (que é castanho) dentro das células da pele horas após a exposição à radiação ultravioleta

d) Sempre que apanha radiações ultravioleta

11) O óleo de lavanda é conhecido pelas suas propriedades…:

a) Energéticas

b) Bronzeadoras

c) Facilitadoras no desaparecimento de celulite

d) Relaxantes

12) “Crest” é uma marca de: a) Shampoos e amaciadores

b) Protectores solares

c) Produtos de higiene dentária

(31)

13) “É importante que o creme protector solar não tenha álcool”…:

a) Porque o álcool actua como factor atenuante do bronzeado

b) Porque quando o álcool entre em contacto com a radiação ultravioleta, aumenta o perigo de sofrermos queimaduras graves na pele

c) Porque o álcool dá um tom mais cinzento à pele

d) Nenhuma das respostas anteriores está certa

14) “Um estado descamativo do couro cabeludo que ocorre como consequência de uma

reprodução anormalmente acelerada das células do couro cabeludo”:

a) É a definição de caspa

b) Acontece quando lavamos o cabelo em demasiadas vezes por dia

c) É a definição de piolhos

d) Nenhuma das restantes respostas está certa

15) Para manter o cabelo no estado mais saudável possível, devemos lavá-lo…: a) 1 vez por semana

b) 3 vezes por semana

c) Todos os dias

Imagem

Table 1: Summary of the average rating per condition

Referências

Documentos relacionados

To instruct the Director of the Pan American Sanitary Bureau to include on the agenda of the XI Meeting of the Directing Council a topic on the problems arising from

With this in view, and on the recommendation of Committee I, the Conference adopted Resolution XXII, instructing the Director to include on the agenda of the XI Meeting of

[Kim et al., 2012] reported that the inhibitory effect of kahweol on cancer metastasis may contribute to inhibition of the expression and secretion of some MMPs

The consumers who reported the correct concept of meat with certification of origin – products which offer information on origin – believe that such certification contributes

ABSTRACT - The objective of this study was to identify the purchase habits and preferences of beef consumers, their level of knowledge on brands and products with

Embora, o PPP (2018) anuncie que a escola reconhece a importância da formação continuada do professor e a equipe diretiva demonstrasse atenção à questão dos

Even though many South African consumers are becoming more price sensitive, footwear imports from Asia have increased and low-cost products are more widely

This paper concludes that Portugal has comparative advantages in products relative intensive in unskilled labour and in products of economic activities with a high level of